Question #12
Reading: Reading 2 Time-Series Analysis
PDF File: Reading 2 Time-Series Analysis.pdf
Page: 5
Status: Unattempted
Correct Answer: A
Part of Context Group: Q11-12
Shared Context
Question
Supposing the time series is actually a random walk, which of the following approaches would be appropriate prior to using an autoregressive model?
Answer Choices:
A. First differencing the time series
B. ARCH
C. Convert the time series by taking a natural log of the series
Explanation
First differencing often transforms a random walk into a covariance stationary time series
which can then be fitted using autoregressive models. ARCH is a type of AR model where
the residuals exhibit conditional heteroskedasticity and is not an approach to convert a
random walk into a covariance stationary time series. Taking natural log is recommended
for a time series with an exponential growth prior to fitting a trend model.